Recognise changes in the "behaviour" of a financial time series (Trading on the news) - page 3
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
write - it's easy to output parameters to a log file via Print... or preferably to a separate file via file operations...
and then also make a table by Exodus when there was ++ and after which number of Minuses...
because it's possible that at least the total number of --- will be greater than +++
but by manipulating lots you can generally output the result in +++
As soon as I started programming in mql, I wrote a quote upload to excel.
You propose to go to differences, if D(Currency)>0, then "+" otherwise "-". then if neighbours "+" and "-", then result variable y(t)=1.
Uh-huh - to the differences... and the +-+-+++-+-+-+-+-+-+-+
to tabulate the outcomes... + -+ -+ + + --+ -+ -+ -+
+ appeared on 1 "turn" - 4 times
+ Appeared on "turn 2" (-+) 5 times
+ Appeared on 3 "turns" ( --+ ) 1 time
---
this makes it easier to see what we can do next with the sample....
Uh-huh - to the differences... and the +-+-+++-+-+-+-+-+-+-+
to tabulate the outcomes... + -+ -+ + + --+ -+ -+ -+
+ appeared on 1 "turn" - 4 times
+ Appeared on "turn 2" (-+) 5 times
+ Appeared on 3 "turns" ( --+ ) 1 time
---
this makes it easier to see what we can do next with the sample....
definitely won't work, essentially a shift to probability, no link to anything.
Recognise changes in the "behaviour" of a financial time series (Trading on the news)
Start by gathering statistics. Start by collecting all the positive news and see how the price behaved before/afterwards. Do the same for negative news. If you find a pattern, you can become an effective "agent".
There is no such connection. Positive news can either cause the market to rise or fall or not move at all. It has been known for a long time.
Of course not. But he must see for himself, otherwise he will continue to be deluded.
s.s. What's to say, even 9/11 had no impact on the dollar, but this is news.
There is no complete solution, only approaches.
The problem is understood intuitively. To fit the model, the bigger the sample, the better. But the bigger the sample, the less it takes into account the current situation. It would seem that AP(1) is ideal - only previous candle, but no, it works, but very rarely.
The model should predict the breaks, but how to do it?
Why predict when it is possible to detect them, and quite early.
Maybe I was unclear, the idea is not to choose the "shortest" model, but to fix the selected "sometimes working" model, but to require that the data corresponds to the model very precisely, much more precisely than usual (say, within 0.1-0.2 sigma). If we have exceeded this narrow interval - we immediately stop trading. Come back - trade again.
Maybe I didn't make myself clear, the idea is not to choose the "shortest" model, but to fix the "sometimes working" model, but to require that the data corresponds to the model very precisely, much more precisely than usual (say, within 0.1-0.2 sigma). If we have exceeded this narrow interval - we immediately stop trading. Come back - trade again.
Of course it didn't. But he must see for himself otherwise he will continue to be deluded.
s.s. What can I say, even 9/11 had no impact on the dollar and this is news.
I posted the result in the Econometrics branch: one step forecast. The accuracy of model fitting within a sample does not affect the accuracy of the prediction. You have to be able to use the forecast, which is a separate problem called "predictability". I've been in the thread, bringing the team up to this problem, but no one understood it.